Assessment and improvement of image homogeneity in black-blood T2-weighted turbo spin-echo CMR
- PDF / 424,038 Bytes
- 2 Pages / 595.28 x 793.7 pts Page_size
- 31 Downloads / 188 Views
ORAL PRESENTATION
Open Access
Assessment and improvement of image homogeneity in black-blood T2-weighted turbo spin-echo CMR Benjamin Wince1*, Lowie M Van Assche1, Han W Kim1, Lubna Bhatti1, Christoph J Jensen1, Elizabeth Jenista1, Wolfgang G Rehwald2, Deneen Spatz1, Yong-Yin Kim1, Michele Parker1, Raymond J Kim1 From 15th Annual SCMR Scientific Sessions Orlando, FL, USA. 2-5 February 2012 Background Double inversion recovery (DIR) and triple inversion recovery (TIR) prepared Turbo-Spin-Echo (TSE) are the most commonly used sequences for black-blood T2weighted (T2W) cardiac magnetic resonance. For both, cardiac motion often leads to signal loss and image inhomogeneity, which can affect diagnosis. Signal loss artifacts are thought to be due to misalignment of the black-blood preparation with readout. However, the TSE readout itself is also motion sensitive and could lead to signal loss. We examined image homogeneity in routine Black-Blood T2W-TSE and investigated potential improvement by reducing interecho-spacing of the TSE readout. Methods Ten healthy volunteers underwent T2W-CMR using 3 sequences: (a) Standard DIR-TSE (interecho-spacing = 9.66ms), (b) Standard TIR-TSE (interecho-spacing = 9.66ms) and, (c) a Modified DIR-TSE sequence with low interecho-spacing (interecho-spacing = 3.48ms). Reduction in interecho-spacing was accomplished by increasing bandwidth >4-fold and utilizing shorter refocusing pulses. Total readout time was held constant for all three sequences, leading to shorter breath-hold times for the modified sequence. In each volunteer, a midventricular short-axis slice was repeatedly imaged with only the timing of readout in diastole changed by 50 ms increments (between 60-100% of the cardiac cycle). All images were acquired using coil normalization, slice thickness=7mm, and effective echo-time=60ms. 1
Cardiology, Duke University, Durham, NC, USA Full list of author information is available at the end of the article
Epicardial and endocardial contours were planimetered. Mean image intensity and standard deviation (SD) were measured for the slice. Image homogeneity was defined as the ratio of SD/mean image intensities x100 (higher values indicate worse homogeneity). The sensitivity of each sequence to readout timing was determined by the maximum change in image homogeneity across the different timepoints of diastole.
Results Overall, mean image homogeneity was best for Modified DIR-TSE (16.1±7.9), compared with Standard DIR-TSE (23.5±11.3) and Standard TIR-TSE (27.1±13.9, both p
Data Loading...
